medRxiv preprint doi: https://doi.org/10.1101/2021.06.07.21258522; this version posted June 12, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC 4.0 International license .

1 SARS-CoV-2 in wastewater from City used for irrigation in the Mezquital 2 Valley: quantification and modelling of geographic dispersion

3

4 Authors

5 Yaxk’in Coronado1, Roberto Navarro2, Carlos Mosqueda2,3, Valeria Valenzuela2,4, Juan 6 Pablo Perez2, Víctor González-Mendoza1, Mayra de la Torre2, Jorge Rocha1*.

7

8 1Conacyt-Unidad Regional , Centro de Investigación en Alimentación y Desarrollo. 9 Ciudad del Conocimiento y la Cultura de Hidalgo, Blvd. Santa Catarina S/N, San Agustín 10 Tlaxiaca, Hidalgo, México, 42163

11 2Unidad Regional Hidalgo. Centro de Investigación en Alimentación y Desarrollo. Ciudad 12 del Conocimiento y la Cultura de Hidalgo, Blvd. Santa Catarina S/N, San Agustín Tlaxiaca, 13 Hidalgo, México, 42163.

14 3Instituto Tecnológico de Celaya. Antonio García Cubas 600, Fovissste, Celaya, Gto., 15 38010.

16 4Universidad Tecnológica de Querétaro. Av. Pie de la Cuesta 2501, Nacional, Santiago de 17 Querétaro, Qro., 76148.

18 *Corresponding author: [email protected]

19 Additional data: 20 https://nbviewer.jupyter.org/github/yaxastro3000/COVID_CIAD_URH/blob/c65ac45af147 21 36023e94eb087aea8d541a7dac68/MODEL_COVID_CIAD_URH.ipynb

22

NOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice. medRxiv preprint doi: https://doi.org/10.1101/2021.06.07.21258522; this version posted June 12, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC 4.0 International license .

23 Abstract

24 Quantification of SARS-CoV-2 in urban wastewaters has emerged as a cheap, efficient 25 strategy to follow trends of active COVID-19 cases in populations. Moreover, 26 mathematical models have been developed that allow prediction of active cases following 27 the temporal patterns of viral loads in wastewaters. In Mexico, no systematic efforts have 28 been reported in the use of this strategies. In this work, we quantified SARS-CoV-2 in 29 rivers and irrigation canals in the , Hidalgo, an agricultural where 30 wastewater from is distributed and used for irrigation. Using quantitative RT- 31 PCR, we detected the virus in 6 out of 8 water samples from rivers, and 5 out of 8 water 32 samples from irrigation canals. Notably, samples showed a general consistent trend of 33 having the highest viral loads in the sites closer to Mexico City, indicating that this is the 34 main source that contributes to detection. Using the data for SARS-CoV-2 concentration in 35 the river samples, we generated a simplified transport model that describes the spatial 36 patterns of dispersion of virus in the river. We suggest that this model can be extrapolated 37 to other wastewater systems that require knowledge of spatial patterns of viral dispersion at 38 a geographic scale. Our work highlights the need for improved practices and policies 39 related to the use of wastewater for irrigation in Mexico and other countries.

40 Introduction

41 The ongoing global pandemic of COVID-19 disease, caused by severe acute respiratory 42 syndrome coronavirus 2 (SARS-CoV-2), is a public health emergency of international 43 concern (Organization and Fund (UNICEF), 2020a, 2020b). SARS-CoV-2 ribonucleic acid 44 (RNA) has been detected in feces from both symptomatic and asymptomatic patients (Chen 45 et al., 2020; Holshue et al., 2020; Jiehao et al., 2020; Tang et al., 2020; W. Wang et al., 46 2020; Zhang et al., 2020) and in wastewater (Ahmed et al., 2020; Lodder and Husman, 47 2020; Medema et al., 2020). For this reason, quantification of SARS-CoV-2 ARN in 48 wastewater has emerged as a cheap, efficient method for monitoring active cases in large 49 populations (Ahmed et al., 2020; S. Wang et al., 2020), small towns (Kitajima et al., 2020; 50 Randazzo et al., 2020), or campuses (Harris-Lovett et al., 2021). Notably, this strategy 51 allows for a one-week anticipation in the active cases, compared to health systems medRxiv preprint doi: https://doi.org/10.1101/2021.06.07.21258522; this version posted June 12, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC 4.0 International license .

52 registries, since asymptomatic individuals contribute to the viral load in wastewaters 53 (Vallejo et al., 2020; Wu et al., 2020).

54 Mexico City has more than 661,446 confirmed cases of COVID-19 (Mexico City 55 Government, 2020; consulted on June 1st, 2021); this is the city with the highest number of 56 cases in the country. In Hidalgo, a state north of Mexican Valley Metropolitan Area 57 (MVMA, Figure 1a), more than 39,012 positive cases have accumulated (State of Hidalgo 58 Government, 2021), of which more than 11,000 cases (28%) correspond to the Mezquital 59 Valley, a highly productive agricultural region.

60 To date, few systematic efforts have been made in Mexico to detect SARS-CoV-2 in 61 wastewater. However, the Mezquital Valley has several relevant characteristics for the 62 study of SARS-CoV-2 in wastewater: 1) agriculture production is maintained by using 63 exclusively wastewater for irrigation (Contreras et al., 2017); 2) the wastewater source is 64 the MVMA, the most populated metropolitan area worldwide, and the region that 65 concentrates most active cases in Mexico (Figure 1b) (Información referente a casos 66 COVID-19 en México - datos.gob.mx/busca); 3) the system includes one of the largest 67 wastewater treatment facilities in , which feeds a river and a complex system 68 of irrigation canals; 4) farmers, inhabitants and consumers in the Mezquital Valley are in 69 contact with water, soil or agricultural products and 5) the use of wastewater for irrigation 70 is one of the main causes that allowed that this Valley is no longer in extreme poverty 71 (García-Salazar and García-Salazar, 2019).

72 Using data for SARS-CoV-2 concentration in wastewater, several models have been 73 proposed for to find a correlation between the temporal patterns of viral concentration and 74 the number of active COVID-19 cases (Hart and Halden, 2020b). Likewise, a previous 75 work presented the first model of spatial and temporal patterns of viral loads in seaway 76 systems, showing the dispersion of the virus in an urban region (Hart and Halden, 2020a). 77 However, models for spatial patterns of viral dispersion are needed in open waterbodies 78 that contain wastewater, in order to understand the virus transport and to identify the zones 79 with higher risk of infection.

80 In this study we sampled water from the River, Salado River and irrigation canals in 81 the Mezquital Valley, which receive wastewater from Mexico City, to assess the presence medRxiv preprint doi: https://doi.org/10.1101/2021.06.07.21258522; this version posted June 12, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC 4.0 International license .

82 of SARS-CoV-2 and generate a mathematical model that describes its spatial dispersion. 83 We propose that our results highly relevant not only to follow the epidemic in Mexico City 84 and municipalities in Hidalgo, but also to evaluate a possible risk of transmission through 85 environmental matrices and measure the stability of SARS-CoV-2 with geographic 86 resolution.

87 Materials and Methods

88 Sampling. Water samples were taken along the between the mouth of the 89 Central Interceptor Tunnel and ; samples were also taken in the Salado River, 90 which receives wastewater from the Grand Drainage Canal, and river, which 91 contains wastewaters from local municipalities (Figure 1). We also collected water samples 92 from irrigation canals in locations representative of the Mezquital Valley. Appropriate 93 safety equipment was always used, consisting of a cotton lab coat, latex gloves, KN95 94 respirator-mask, rubber boots, disposable cap, and safety glasses. A simple sampling 95 technique was used (NOM, 1980; EPA, 2017), locating sites where the wastewater is well 96 mixed near to the center of the flow channel, approximately between 40 and 60 percent of 97 water depth, where turbulence is maximum, and there is minimum sedimentation of solids. 98 For each location, three water samples were collected as follows: two samples of 400 ml in 99 glass bottles, for SARS-CoV-2 detection and for microbiological analyses; and one 4 L 100 sample in a plastic bottle, for physicochemical analysis. For each water sample collected at 101 an irrigation canal, we sampled 500 g of soil from an adjacent agricultural field. All 102 samples were immediately placed in ice until arrival at the laboratory. Water samples for 103 SARS-CoV-2 detection were inactivated by incubation at 60˚C for 1 h upon arrival at the 104 laboratory.

105 Physicochemical and Microbiological Analyses. Samples for physicochemical and 106 microbiological analyses were stored in ice from sampling until delivery to external 107 laboratories where determinations were carried out. Due to the remoteness of sampling 108 sites, analyses were performed between 18 to 40 h after sampling. Water samples were 109 analyzed for physicochemical parameters at Laboratorio de Análisis Químicos Analíticos 110 (LAQA), at Facultad de Química, Universidad Autónoma de Querétaro (Querétaro, 111 México). Physicochemical parameters measured were: Biochemical oxygen demand medRxiv preprint doi: https://doi.org/10.1101/2021.06.07.21258522; this version posted June 12, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC 4.0 International license .

112 (BOD), chemical oxygen demand (COD); pH, total dissolved solids (TDS), suspended 113 solids (SS), total suspended solids (TSS) and total solids (TS). Microbiological analysis of 114 total coliforms (TC), fecal coliforms (FC), Escherichia coli and Salmonella (PCR 115 detection) were analyzed (according to norm NOM-210-SSA-2014) at Laboratorio para la 116 Evaluación y Control de Riesgos Microbianos en Alimentos (LECRIMA) at UAQ 117 (Querétaro, México). The ‘most probable number’ method was used for quantification of 118 TC, FC and E. coli.

119 Viral RNA Extraction. For soil samples, viral particles were first eluted from 15 g of sample 120 in 150 ml of 10% beef extract buffer (pH 7) by stirring for 30 min (Horm et al., 2012). 121 Next, soil suspension and inactivated wastewater samples (150 ml) were filtered through a 122 No. 5 Whatman filter (Millipore), and then through 0.22 µm polyethersulfone membrane 123 (Corning). Next, viral particles were precipitated by adding NaCl (100 g/L) and 124 Polyethilenglicol (22 g/L) and stirring for 20 min. Then, 105 ml samples were centrifuged 125 at 12,000 x g for 2 hours, and the resulting pellet was suspended in 400 µl RNAse free 126 water. Viral nucleic acids were extracted from these samples using the Purelink Viral 127 DNA/RNA kit (Invitrogen). Purified nucleic acid samples were quantified by 128 spectrophotometry in a Nanodrop One (Thermo Scientific).

129 Quantification of SARS-CoV-2 RNA. For detection of SARS-CoV-2, the kit Decov2 Triplex 130 (Genes2Life, Irapuato, Mexico) was used, which includes reverse transcription and PCR 131 detection of three targets of N ORF (N1-FAM, N2-HEX and N3-TexasRed) in a single 132 multiplex reaction. For each reaction, 5 µl of viral RNA were added as template in a total 133 volume of 25 µl. The RT-qPCR reaction was performed in a QuantStudio 5 (Thermo 134 Scientific) instrument, and settings for passive reference was changed to ‘none’. Duplicate 135 SARS-CoV-2 detection reactions were performed for each RNA. For absolute 136 quantification of SARS-CoV-2 in environmental samples, a standard curve was obtained 137 using synthetic RNA targets provided with the detection kit. For all RNA samples, we 138 included an internal control reaction for the detection of Pepper Mild Mottle Virus 139 (PMMV), using the primers Fwd 5’- GAG TGG TTT GAC CTT AAC GTT TGA-3’, Rev 140 5’- TTG TCG GTT GCA ATG CAA GT-3’ and probe 5’-FAM- CCT ACC GAA GCA 141 AAT G-BHQ1-3 (Zhang et al., 2005; Haramoto et al., 2020). medRxiv preprint doi: https://doi.org/10.1101/2021.06.07.21258522; this version posted June 12, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC 4.0 International license .

142 Modelling of water dispersion of SARS-CoV-2 in the Tula River. The transport model for 143 dispersion of virus in unsaturated media (Torkzaban et al., 2006) was adapted. The original 144 model describes the viral concentration with respect to the distance from the virus input site 145 and the decrease over time, and considers terms of adsorption and detachment of virus with 146 respect to the soil-water interface and air-water interface. We considered only dispersion in 147 the water bulk; we also considered that dispersion of virus is independent on time, and it 148 depends on the distance. Hence, the attachment and detachment rate coefficients and the a a s 149 inactivation rate coefficient in soil and air media were equal to zero (κ att = 0, κ det = 0, κ att s 150 = 0, κ det = 0, μs = 0 and μa = 0). We also assume that Darcy flux (q) and inactivation rate

151 coefficient (μw) are constant along the river. The variables amount of water (θ) and 152 coefficient of dispersion (D) were fitted to the experimental data. Finally, we excluded the 153 contributions of the adsorption from the general model and considered two entrances that 154 affect the viral concentration. Detailed derivation of the equations is shown in 155 Supplementary File 1. The resulting equation (1) is shown below:

∂θCw ∂ ∂Cw ∂qCw = θD − − μwθCw (1) ∂t ∂x ( ∂ x ) ∂ x

156 Where Cw, is the viral concentration and the other variables are previously decribed.

157 The equation was solved numerically with initial and boundary conditions using the 158 software package Scipy Odeint in Python. The concentration values for the virus in the 159 Tula River were obtained experimentally as described above.

160 Results

161 Social, agronomic and economic considerations of the study system

162 The Mezquital Valley is located 80 km north of Mexico City (Fig. 1). It is inhabited mainly 163 by the indigenous Otomí group that speaks a native language called hñähñu. The average 164 annual precipitation is 409 mm (Moreno Alcántara, Garret Ríos and Fierro Alonso, 2006) 165 and the main economic activity is agriculture, but the arid climate contributes to the 166 extreme poverty, historically associated to the region (SAGARPA, 2003; Instituto Nacional 167 de Lenguas Indígenas (Mexico), 2009; Rossette, 2017). medRxiv preprint doi: https://doi.org/10.1101/2021.06.07.21258522; this version posted June 12, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC 4.0 International license .

168

169 Figure 1. Study area. a) Location of the State of Hidalgo, in Mexico. Dashed area corresponds to Figure 170 1B. b) Mexico City is the main source of wastewater used for irrigation in the Mezquital Valley. 171 HeatMap showing the population density in the study area. Yellow, blue and green arrows show the 172 location and discharge of the main sewage systems. WWTP, Wastewater treatment plant. Dashed area 173 corresponds to Figure 1c. c) Sampling locations and dates for samples in the Mezquital Valley. For 174 agricultural field locations, samples of both water (irrigation canal) and soil (adjacent agricultural field) 175 were collected.

176

177 Wastewater is received from Mexico City through three canals: West Interceptor Tunnel, 178 Central Interceptor Tunnel, and Grand Drainage Canal (Fig. 1b). Subsequently, the water is 179 collected in the Irrigation Districts (003-Tula, 100-, and 112-), and a medRxiv preprint doi: https://doi.org/10.1101/2021.06.07.21258522; this version posted June 12, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC 4.0 International license .

180 dam system that includes Taxhimay, Requena, Endhó, Javier Rojo Gómez, and Vicente 181 dams. This dam system is considered the largest in the world (Islas, 2010). 182 Wastewater is partially processed in the wastewater treatment plant (WWTP) in Atotonilco 183 de Tula, Hidalgo (Fig. 1c), with a maximum treatment capacity of 50 m3/s. In Hidalgo, five 184 Irrigation Districts distribute 1,513,395,400 m3/year of water (CONAGUA, 2019); 97% 185 corresponds to the irrigation districts 003 of Tula, 100 Alfajayucan, and 112 Ajacuba, 186 which provide residual water to the agricultural sector of the Mezquital Valley and belong 187 to the hydrological-administrative region XIII Mexico Valley.

188 The WWTP normally works at 30% of its capacity because the producers refuse to use the 189 treated water in their irrigation systems. For this reason, this is a health and environmental 190 risk because these wastewaters irrigate more than 80,000 ha of crops (Lesser et al., 2018). 191 However, the use of wastewater from Mexico City has had a positive impact on the 192 productivity of the diverse crops grown (Institute for Federalism and Municipal 193 Development (INAFED), 2010). Wastewaters contribute with large amounts of organic 194 matter during decreasing the costs of fertilization. These factors have allowed an increase in 195 land rentability, reaching >1,000 USD per ha per year. In turn, wastewater treatment could 196 directly increase production costs, price of water, and increased needs for fertilizers, which 197 could impact land rentability (Pérez Camarillo, 2002; Pérez, Zacatenco and Martínez, 2006; 198 SIAP, 2018).

199

200 Sample collection

201 Between September 22nd and October 6th 2020, we collected water samples in eight 202 locations of the Tula River, which represent the main flow of wastewater from Mexico City 203 in the Mezquital Valley, this also included two locations in the Tepeji River and Salado 204 River (Fig. 1c, Table 1). Water samples were also collected from irrigation canals adjacent 205 to agricultural fields as well as soil samples from these agricultural fields (Fig. 1c, Table 1). 206 Produce samples were also collected at three of the agricultural fields: a sample of 207 coriander at site 3, and samples of lettuce at sites 6 and 7. In total, our sampling covered 208 ≈80 km in the Mezquital Valley, and included the municipalities of Tepeji del Río, 209 , , de Aldama, de Juárez, medRxiv preprint doi: https://doi.org/10.1101/2021.06.07.21258522; this version posted June 12, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC 4.0 International license .

210 , and Ixmiquilpan (Table 1).

211 Table 1. Water samples collected in Mezquital Valley for quantification of SARS-CoV-2.

Code Location Waterbody Latitude Lenght Description

A mix of water-springs and municipal discharge from Tepeji-El Salto RW1 Tepeji River 19.91062 -99.33189 localities of Tepeji del Río de Bridge Ocampo flows into the Requena Dam, which feeds the Tula River.

Receives wastewater discharges from the Mexico City Central Metropolitan Area (MCMA). RW2 Interceptor Tula River 19.953149 -99.297094 These are discharged into “El Tunnel Mouth Salto” river, which becomes “Tula” river from this point.

Cooperativa Intermediate location between RW3 Cruz Azul City Tula River 19.987822 -99.323474 Emisor Central and the city of Bridge Tula de Allente.

Tula de Allende Intermediate location between RW4 Tula River 20.05919 -99.340291 City Bridge Requena Dam and Endhó Dam

Intermediate location between Tezontepec RW5 Tula River 20.194539 -99.281055 Endhó Dam and the convergence City Bridge of Salado river and Tula river.

Location downstream of the Mixquiahuala RW6 Tula River 20.240323 -99.230846 convergence of Salado river and City Bridge Tula river.

Salado river; accumulates municipal wastewater and RW7 Mangas Bridge Salado River 20.184949 -99.25571 receives wastewater from MCMA Gran Canal de Desague

RW8 Ixmiquilpan Tula River 20.483365 -99.22058 Tula river at Ixmiquilpan; this medRxiv preprint doi: https://doi.org/10.1101/2021.06.07.21258522; this version posted June 12, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC 4.0 International license .

City Bridge location if 80 km downstream of Emisor Central

Water from the Wastewater Treatment Plant (WWTP). In the literature, the WWTP registers CW1 Requena Canal Final Canal 20.191971 -99.243117 that it treats only 30% of all the flow from the MCMA. However, the CONAGUA assures that it currently treats 40 to 60%.

Water from the Endhó Dam. Intermediate Once the Tula River leaves the CW2 Endho Canal 20.215935 -99.203465 Canal city of Tula de Allende, it reaches the Endhó Dam.

Water from the Wastewater CW3 Requena Canal Main Canal 20.156447 -99.234083 Treatment Plant.

According to producers, the WWTP only treats less than 50% of the Central Interceptor Tunnel CW4 WWTP Canal Main Canal 19.963483 -99.301255 discharge effluent. The treated water is mixed with residual water, and the flow continues through the irrigation canals.

It is also called Tlamaco-Juandhó. Fuerza Canal It carries a mixture of water from CW5 (Tlamaco- Final Canal 20.087938 -99.2049 municipal discharges and Juandho) wastewater from the WWTP, Tequisquiapan, and Zumpango.

The endpoint of irrigation canals,

Alto-Ajacuba coming from the WWTP and a CW6 Final Canal 20.105037 -99.146827 Canal mixture of municipal discharges and thermal-watering places. This area already belongs to the medRxiv preprint doi: https://doi.org/10.1101/2021.06.07.21258522; this version posted June 12, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC 4.0 International license .

Irrigation District 112-Ajacuba.

Water from the Endhó Dam, CW7 Alto Canal Final Canal 20.521108 -99.133735 passing through the Tula River.

Water from the slopes of the Intermediate CW8 Xotho Canal 20.526834 -99.1095 Tepatepec, Tezontepec, Canal Mixquiahuala Canals.

212 RW; water sample from Tula River. CW; water sample from irrigation canal.

213

214 SARS-CoV-2 quantification in water samples from Tula River

215 We detected RNA from SARS-CoV-2 in water samples collected in the Tula River, Tepeji 216 River and Salado Rover (Fig. 2). Detection was achieved from the three targets included in 217 the kit (N1-FAM, N2-HEX and N3-TexasRed) in most cases; however, quantification using 218 N2-HEX Ct values and its corresponding standard curve yielded more consistent results, 219 and absolute values were always intermediate to those obtained with N1-FAM and N3- 220 TexasRed (not shown). From these criteria, we use N2-HEX data throughout this study. 221 SARS-CoV-2 RNA was detected in 6 out of 8 water samples from river (Fig. 2). As 222 expected, the highest concentration was found in sample RW2, at the entrance flow of the 223 wastewater treatment plant (WWTP, Fig. 1c and 2), with a concentration of 79 RNA copies 224 per ml; from the sampled locations in the Tula River, this is the closest to the Mexican 225 Valley Metropolitan Area (MVMA, Fig. 1 and 2). This value decreased through the flow of 226 the river (which consists mostly of untreated wastewater) in samples RW3 and RW4, and 227 was undetectable in RW5, downstream of the Endhó Dam. SARS-CoV-2 was also detected 228 at the entrance fluxes of Tepeji River (RW1, 18 copies per ml) and Salado River (RW6, 24 229 copies per ml), both of which contribute to viral load in the Tula River. Finally, SARS- 230 CoV-2 was not the detected in sample RW8, which was collected at Ixmiquilpan, ≈40 km 231 downstream of sample RW7 (Fig. 2). medRxiv preprint doi: https://doi.org/10.1101/2021.06.07.21258522; this version posted June 12, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC 4.0 International license .

232 233 Figure 2. Quantification of SARS-CoV-2 ARN in water samples from the Tula River. a) Schematic 234 representation of distances and fluxes present between the sampling locations at the Tula, Tepeji and 235 Salado River. WWTP, Wastewater treatment plant. b) RT-qPCR results of SARS-CoV-2 quantification 236 in the Tula River.

237

238 SARS CoV-2 RNA in irrigation canals.

239 RNA from SARS-CoV-2 was also detected in water samples collected in irrigation canals 240 throughout the Mezquital Valley. Irrigation canals are mostly fed from treated water, 241 however, untreated wastewaters from municipalities in Hidalgo are also connected to these 242 canals. Since the irrigation canal network in the Mezquital Valley is highly complex and a medRxiv preprint doi: https://doi.org/10.1101/2021.06.07.21258522; this version posted June 12, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC 4.0 International license .

243 detailed map is unavailable, we only show the general known source of water in each 244 location (Fig. 3a). RNA From SARS-CoV-2 was detected in 5 out of 8 water samples 245 collected from irrigation canals (Fig. 3b). The highest concentration of 112 copies per ml 246 was found in sample CW1 (Fig. 3b). This location is the closest from MVMA, however, it 247 is reportedly fed solely from wastewater treated in the WWTP (Fig. 3a, table 1). SARS- 248 CoV-2 ARN was also detected in samples CW2, CW4, CW4-II and CW5, all of which are 249 fed from the WWTP, but also from wastewater from municipalities in Hidalgo, and from 250 the Endhó Dam (Fig. 3). SARS-CoV-2 was not detected in samples CW3, CW6 and CW7; 251 notably, irrigation canals at Ixmiquilpan (CW6 and CW7) are fed from freshwater bodies 252 (not wastewater) such as groundwater wells or local spas, while CW3 is reportedly fed 253 from the WWTP and local municipalities (Fig. 3).

254 Since SARS-CoV-2 RNA was detected in water samples from irrigation canals, we 255 hypothesized that it may be detectable in agricultural soil from adjacent fields (Fig. 1c, 256 Table 1). However, SARS-CoV-2 was not detected in these samples (Supplementary Table 257 S1 and S2). Similarly, SARS-CoV-2 RNA was not detected in any of the produce samples, 258 which were collected in the field or at the Ixmiquilpan Market (Table 1, Supplementary 259 Table S2). Since the internal control PMMV was detected in all of these samples 260 (Supplementary Table S2), we discard that the presence of PCR inhibitors as the reason for 261 lack of detection, but further studies are needed to confirm the absence of SARS-CoV-2 262 since viral particles can adsorb to soils and other solids.

263 medRxiv preprint doi: https://doi.org/10.1101/2021.06.07.21258522; this version posted June 12, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC 4.0 International license .

264

265 Figure 3. Quantification of SARS-CoV-2 ARN in water from irrigation canals in the Mezquital Valley. 266 a) Schematic representation showing the reported source of water in irrigation canals sampled. b) RT- 267 qPCR results of SARS-CoV-2 quantification in water samples from irrigation canals.

268

269 medRxiv preprint doi: https://doi.org/10.1101/2021.06.07.21258522; this version posted June 12, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC 4.0 International license .

270 Physicochemical and microbiological analyses of water samples

271 Results on physicochemical analyses are shown in Supplementary Table S3 and results on 272 microbiological analyses are shown on Supplementary Table S4. Based on standards 273 stablished by Mexican laws, all water samples had values of BOD, TSS and SS below the 274 permitted limits for water used for irrigation (NOM-001-SEMARNAT-1996). Regarding 275 limits for microbiological parameters (NOM-001-ECOL-1996), 5 out of 8 water samples 276 from the Tula River, and 5 out of 8 water samples from irrigation canals, presented total 277 coliform values above the allowed limits for wastewater used for irrigation (Supplementary 278 Table S4).

279 We assessed if the physicochemical characteristics of wastewater, organic matter and 280 microbial concentration correlated with the concentration of the virus. In samples from the 281 river water, correlation analyses showed that several variables were associated with SARS- 282 CoV-2 RNA concentration. In contrast, analyses using parameters obtained from irrigation 283 canals, showed no significant correlations between SARS-CoV-2 RNA concentration and 284 physicochemical or microbiological variables (Table 2, right). Significant p values were 285 obtained for correlation analyses with physicochemical parameters BOD, COD, pH, and SS 286 (Table 2, left), and also for all three microbiological parameters in river samples (Table 2, 287 Supplementary Figure S1). Further inspection of correlation trends revealed correlations 288 with pH and SS could be spurious, since they were highly grouped in discrete values 289 (Supplementary Figure S1), and a higher number of samples should be analyzed to confirm 290 this observations (Table 1, Suplementary Figure S1).

291

292 Table 2. Correlation of SARS-CoV-2 quantification versus physicochemical and 293 microbiological parameters in water samples.

Physicochemical parameters

River water samples Irrigation canal samples Parameter r2 p value Significance r2 p value Significance

BOD 0.66451416 0.00872825 ** 0.11417103 0.40483251 - medRxiv preprint doi: https://doi.org/10.1101/2021.06.07.21258522; this version posted June 12, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC 4.0 International license .

COD 0.94529181 7.42E-06 *** 0.0181819 0.74745041 -

pH 0.68970848 0.0064762 ** 0.42338299 0.06905202 -

TDS 0.24418989 0.20131137 - 0.3650173 0.10036602 -

SS 0.86636172 0.00024926 *** 0.00302837 0.89594555 -

TSS 0.26441269 0.18013422 - 0.01181227 0.79562296 -

TS 0.17469837 0.29241941 - 0.35416452 0.10724613 -

Microbiological parameters

River water samples Irrigation canal samples Parameter r2 p value Significance r2 p value Significance

TC 0.82551851 0.00070422 *** 0.21502632 0.23573855 -

FC 0.80436747 0.00109741 *** 0.35713184 0.10532899 -

E. coli 0.80436747 0.00109741 *** 0.40315677 0.07888845 -

294 BOD, Biochemical oxygen demand; COD, chemical oxygen demand; pH, Hydrogen potential, TDS, 295 total dissolved solids; SS, suspended solids, total suspended solids; TSS, total suspended solids; TS, total 296 solids; TC, total coliforms; FC, fecal coliforms; **, p<0.01; ***, p<0.005.

297

298 COD and BOD are a measurement of oxygen required for chemical oxidation of organic 299 material in water, and the Biodegradability index (COD/BOD ratio) is an indicator of the 300 non-biodegradable organic-chemicals. We assesed if the Biodegradability index correlates 301 with the virus concentration of the water system (Abdalla and Hammam, 2014). 302 Interestingly, the trend is the same than the virus concentration (Supplementary Figure S2).

303

304 Modelling the geographic dispersion of SARS-CoV-2 in the Mezquital Valley.

305 Using data for SARS-CoV-2 concentration in the Tula River, we sought to adapt a model 306 that describes viral dispersion in the wastewater bulk. Our model only considers the 307 dispersion of the virus in water and the modification of the flux in the mainstream. The medRxiv preprint doi: https://doi.org/10.1101/2021.06.07.21258522; this version posted June 12, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC 4.0 International license .

308 dispersion coefficient was calculated through least square optimization to the experimental 309 data (Suplementary File 1) (Molugaram et al., 2017). The concentration in the initial 310 sampling site RW2 (Figure 1c.) was normalized to 1 (Ratio of Concentration C/initial 311 Concentration Ci); since it was the maximun value, the experimental data were normalized 312 to this value. Additionally, we modeled two entrances at the points RW2 and RW6, as a 313 result the dispersion coefficient increases at these points. The model fitted to the 314 experimental data along the Tula River (Figure 4a). Therefore, with only three variables, 315 dispersion coefficient (D), amount of water (Θ) and a distance function of D [D(x)], we 316 could define the dispersion of the virus in the Tula River. Finally, we used QGIS3 to obtain 317 a geographic representation of our model and the experimental data (Figure 4b), which 318 show the general trends of SARS-CoV-2 dispersion in the Tula River.

319 medRxiv preprint doi: https://doi.org/10.1101/2021.06.07.21258522; this version posted June 12, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC 4.0 International license .

320

321 Figure 4. The transport model generated for SARS-CoV-2 dispersion in the Tula River, fits the data 322 obtained experimentally. a) Graphic representation of viral concentration in the Tula River, comparing 323 the experimental data and the transport model obtained in this study. b) Map representation of the 324 geographic dispersion of SARS-COV-2 presence in the Tula River, comparing the transport model (left) 325 and the experimental data (right).

326 medRxiv preprint doi: https://doi.org/10.1101/2021.06.07.21258522; this version posted June 12, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC 4.0 International license .

327 Discussion

328 In this work, we quantified SARS-CoV-2 ARN in wastewaters from Mexico City, which 329 are distributed throughout the Tula River, Salado River and irrigation canals and are used 330 for irrigation of crops in the Mezquital Valley. Notably, we detected the virus in 331 geographically representative water samples from both rivers and the irrigation canals. 332 With our data, we generated a viral dispersion model which can be used to assess 333 concentration of SARS-CoV-2 to other river systems if the dispersion coefficient (D), 334 amount of water (Θ) and a distance function of D [D(x)] of the river are fitted.

335 Samples from both the Tula River and irrigation canals followed the same general trend of 336 presenting the highest concentration in the sites located near the WWTP. Viral 337 concentration decreased in the remaining sites and was undetectable in the northernmost 338 sites, located near the town of Ixmiquilpan. This result indicates that the most important 339 source of SARS-CoV-2 ARN in these water systems is the wastewater from Mexican City 340 Valley, and that the viral particles tend to degrade or dilute along the flow of water through 341 the Mezquital Valley. However, we do not discard that municipalities in the Mezquital 342 Valley contribute to viral load and fecal contamination of Tula River, Salado River and the 343 irrigation canals.

344 Our results showed that the SARS-CoV-2 virus was not detected on soil samples. While 345 this could be related to PCR inhibitors in the samples, we found that the PMMV virus was 346 detected and hence discarded this hypothesis. Virus associate with soil and other particulate 347 matter and both adsorption and detachment of virus depends on temperature, moisture, pH, 348 and the physicochemical characteristics of the virus capsid surface and the particles. 349 Similarly, the microbes and organic matter may have a protective effect on virus 350 (AzadpourKeeley, Faulkner and Chen, 2003). Further studies are needed to asses if lack of 351 detection of SARS-CoV-2 is related to degradation or dispersion of the virus, or 352 alternatively, if viral particles adsorb in the soil matrix, which would make our ARN 353 extraction method not appropriate.

354 We found that physicochemical variables COD and BOD and microbiological variables 355 showed significant correlations with SARS-CoV-2 quantification, but only in samples from 356 the Tula River. However, these observations should be confirmed with measurements in a medRxiv preprint doi: https://doi.org/10.1101/2021.06.07.21258522; this version posted June 12, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC 4.0 International license .

357 higher number of samples. This result suggests that in the river, anthropological sources 358 from a single source of contamination predominate, i.e. the MVMA. However, in irrigation 359 canals – where no correlations were found – may receive other sources of contamination 360 (from agriculture, livestock, or industry) (Guédron et al., 2014; Lesser et al., 2018). 361 Furthermore, our correlation analyses suggest that BOD, COD, and the Biodegradability 362 Indec (BOD/COD), as well as fecal coliforms may be good indicators of SARS-CoV-2 363 ARN concentration in municipal wastewaters.

364 The model proposed to explain the dispersion of SARS-CoV-2 virus along the Tula River, 365 as well as the experimental data, indicate that the viral concentration decreases from south 366 to north, with an initial dispersion point at the entrance of the Tula River at the WWTP and 367 a minimal contribution from the communities along the Mezquital Valley. Our model 368 predicts viral dispersion in the Mezquital Valley, and may be useful for selection sampling 369 locations prior to temporal monitoring to quantify the evolution of SARS-CoV-2 (or other 370 virus) epidemics in populations. Previously available models for viral dispersion in 371 environmental matrices (Bivins et al., 2020; Farkas et al., 2020; Kitajima et al., 2020) are 372 useful for detailed dissection of variables governing dispersion patterns of viral particles in 373 at a smaller scale in controlled environments. Our model is, to our knowledge, the first to 374 allow prediction of viral dispersion in linear water bodies at a geographic scale. This model 375 could be extrapolated to other rivers, streams, and canals.

376 One of the main purposes of this work was to assess the possible risk of environmental 377 transmission of SARS-CoV-2 through environmental matrices. Indeed, some recent reports 378 suggest/show that the virus may be infective (Kitajima et al., 2020), and hence, direct 379 contact with water bodies in the Mezquital Valley may represent a risk of transmission, 380 especially for farmers during flooding irrigation (Lüneberg et al., 2018). However, this 381 work does not seek to eliminate wastewater usage in the Mezquital Valley, as this practice 382 is one of the key factors that allowed the economic development of the region. In fact, the 383 Mezquital Valley is considered a study model for the agricultural development of arid and 384 semi-arid rural (Anderson, 2020; Bonvehi Rosich and Seth Denizen, 2021; Seth 385 Denizen, 2021). We suggest that our work should be considered for the creation and 386 modification of practices and policies related to water treatment (Belhadi et al., 2020), medRxiv preprint doi: https://doi.org/10.1101/2021.06.07.21258522; this version posted June 12, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC 4.0 International license .

387 pathogen monitoring, agricultural practices, and safety measures for farmers in the 388 Mezquital Valley.

389

390 5. Acknowledgements

391 This work was supported by CONACyT, Mexico, grant 312014 for JR. This work was 392 possible thanks to the collaboration of local farmers in the state of Hidalgo, who allowed us 393 to collect samples in their fields. The authors thank Gabriela Rivas (UPFIM) and Ana 394 Alcalá (IBT UNAM), Laura Palomares (IBT), Oscar Monroy (UAMI) and Jesús Hernández 395 (CIAD Hermosillo) for advice and technical assistance. We thank personnel from 396 governmental institutions in the state of Hidalgo: Jaime Ortega (SEDAGROH), Sergio 397 Guzman (CONAGUA) and José Alonso Huerta (CITNOVA), for their valuable 398 contribution in the stages of planning, sampling, and discussion of results.

399

400 6. References

Abdalla, K. Z. and Hammam, G. (2014) ‘Correlation between Biochemical Oxygen Demand and Chemical Oxygen Demand for Various Wastewater Treatment Plants in Egypt to Obtain the Biodegradability Indices’, International Journal of Sciences, 13(1), p. 7.

Ahmed, W. et al. (2020) ‘First confirmed detection of SARS-CoV-2 in untreated wastewater in : A proof of concept for the wastewater surveillance of COVID-19 in the community’, Science of The Total Environment, 728, p. 138764. doi: 10.1016/j.scitotenv.2020.138764.

Anderson, A. (2020) The Right to Sewage: Agriculture, climate change, and the growing need for cities to embrace wastewater reuse, Harvard Graduate School of Design. Available at: https://www.gsd.harvard.edu/2020/09/the-right-to-sewage-agriculture- climate-change-and-the-growing-need-for-cities-to-embrace-wastewater-reuse/ (Accessed: 28 May 2021).

Belhadi, A. et al. (2020) ‘Infectious Waste Management Strategy during COVID-19 medRxiv preprint doi: https://doi.org/10.1101/2021.06.07.21258522; this version posted June 12, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC 4.0 International license .

Pandemic in : an Integrated Decision-Making Framework for Selecting Sustainable Technologies’, Environmental Management, 66(6), pp. 1085–1104. doi: 10.1007/s00267- 020-01375-5.

AzadpourKeeley, A., Faulkner, P. and Chen, J. (2003) ‘MOVEMENT AND LONGEVITY OF VIRUSES IN THE SUBSURFACE’, Environmental Protection Agency, Washington, DC, EPA, EPA/540/S-03/500, p. 25.

Bivins, A. et al. (2020) ‘Wastewater-Based Epidemiology: Global Collaborative to Maximize Contributions in the Fight Against COVID-19’, Environmental Science & Technology, 54(13), pp. 7754–7757. doi: 10.1021/acs.est.0c02388.

Bonvehi Rosich, M. and Seth Denizen, K. (2021) New Visions for Wastewater Equity in The Mezquital Valley, MCI. Available at: https://research.gsd.harvard.edu/mci/gsd- professors-and-students-propose-new-visions-for-wastewater-equity-in-the-mezquital- valley/ (Accessed: 28 May 2021).

Chen, Y. et al. (2020) ‘The presence of SARS-CoV-2 RNA in the feces of COVID-19 patients’, Journal of Medical Virology, 92(7), pp. 833–840. doi: https://doi.org/10.1002/jmv.25825.

CONAGUA, C. N. del A. (2019) Estadísticas Agrícolas de los Distritos de Riego. Año Agrícola 2017-2018. Mexico: SEMARNAT-CONAGUA. Available at: https://files.conagua.gob.mx/conagua/publicaciones/Publicaciones/EAUR_2017-2018.pdf.

Contreras, J. D. et al. (2017) ‘Health risks from exposure to untreated wastewater used for irrigation in the Mezquital Valley, Mexico: A 25-year update’, Water Research, 123, pp. 834–850. doi: 10.1016/j.watres.2017.06.058.

Farkas, K. et al. (2020) ‘Wastewater and public health: the potential of wastewater surveillance for monitoring COVID-19’, Current Opinion in Environmental Science & Health, 17, pp. 14–20. doi: 10.1016/j.coesh.2020.06.001.

García-Salazar, E. M. and García-Salazar, E. M. (2019) ‘El agua residual como generadora del espacio de la actividad agrícola en el Valle del Mezquital, Hidalgo, México’, Estudios medRxiv preprint doi: https://doi.org/10.1101/2021.06.07.21258522; this version posted June 12, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC 4.0 International license .

sociales. Revista de alimentación contemporánea y desarrollo regional, 29(54). doi: 10.24836/es.v29i54.741.

Harris-Lovett, S. et al. (2021) ‘Wastewater Surveillance for SARS-CoV-2 on College Campuses: Initial Efforts, Lessons Learned, and Research Needs’, International Journal of Environmental Research and Public Health, 18(9), p. 4455. doi: 10.3390/ijerph18094455.

Hart, O. E. and Halden, R. U. (2020a) ‘Modeling wastewater temperature and attenuation of sewage-borne biomarkers globally’, Water Research, 172, p. 115473. doi: 10.1016/j.watres.2020.115473.

Hart, O. E. and Halden, R. U. (2020b) ‘Simulated 2017 nationwide sampling at 13,940 major U.S. sewage treatment plants to assess seasonal population bias in wastewater-based epidemiology’, The Science of the Total Environment, 727, p. 138406. doi: 10.1016/j.scitotenv.2020.138406.

Holshue, M. L. et al. (2020) ‘First Case of 2019 Novel Coronavirus in the United States’, New England Journal of Medicine, 382(10), pp. 929–936. doi: 10.1056/NEJMoa2001191.

Horm, V. S. et al. (2012) ‘Highly Pathogenic Influenza A(H5N1) Virus Survival in Complex Artificial Aquatic Biotopes’, PLOS ONE, 7(4), p. e34160. doi: 10.1371/journal.pone.0034160.

Institute for Federalism and Municipal Development (INAFED) (2010) Regiones Naturales. Encyclopedia of the Municipalities and Delegations of Mexico. Ministry of the Interior of Mexico. Institute for Federalism and Municipal Development (INAFED). Available at: http://www.inafed.gob.mx/work/enciclopedia/EMM13hidalgo/regionalizacion.html.

Instituto Nacional de Lenguas Indígenas (Mexico) (ed.) (2009) Catálogo de las lenguas indígenas nacionales: variantes lingüísticas de México con sus autodenominaciones y referencias geoestadísticas. 1. ed. México, D.F: Instituto Nacional de Lenguas Indígenas.

Islas, O. (2010) Resultados del estudio de Diagnóstico Sectorial en el Estado de Hidalgo PDF Free Download. Available at: https://docplayer.es/10681820-Resultados-del-estudio- medRxiv preprint doi: https://doi.org/10.1101/2021.06.07.21258522; this version posted June 12, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC 4.0 International license .

de-diagnostico-sectorial-en-el-estado-de-hidalgo-2010.html (Accessed: 28 May 2021).

Jiehao, C. et al. (2020) ‘A Case Series of Children With 2019 Novel Coronavirus Infection: Clinical and Epidemiological Features’, Clinical Infectious Diseases, 71(6), pp. 1547– 1551. doi: 10.1093/cid/ciaa198.

Kitajima, M. et al. (2020) ‘SARS-CoV-2 in wastewater: State of the knowledge and research needs’, Science of The Total Environment, 739, p. 139076. doi: 10.1016/j.scitotenv.2020.139076.

Lesser, L. E. et al. (2018) ‘Survey of 218 organic contaminants in groundwater derived from the world’s largest untreated wastewater irrigation system: Mezquital Valley, Mexico’, Chemosphere, 198, pp. 510–521. doi: 10.1016/j.chemosphere.2018.01.154.

Lodder, W. and Husman, A. M. de R. (2020) ‘SARS-CoV-2 in wastewater: potential health risk, but also data source’, The Lancet Gastroenterology & Hepatology, 5(6), pp. 533–534. doi: 10.1016/S2468-1253(20)30087-X.

Lüneberg, K. et al. (2018) ‘Drylands soil bacterial community is affected by land use change and different irrigation practices in the Mezquital Valley, Mexico’, Scientific Reports, 8(1), p. 1413. doi: 10.1038/s41598-018-19743-x.

Medema, G. et al. (2020) ‘Presence of SARS-Coronavirus-2 RNA in Sewage and Correlation with Reported COVID-19 Prevalence in the Early Stage of the Epidemic in The Netherlands’, Environmental Science & Technology Letters, 7(7), pp. 511–516. doi: 10.1021/acs.estlett.0c00357.

Mexico City Government (2021) Available at: https://datamexico.org/es/coronavirus.

Molugaram, K. et al. (2017) Statistical Techniques for Transportation Engineering. Butterworth-Heinemann.

Moreno Alcántara, B., Garret Ríos, M. G. and Fierro Alonso, U. (2006) Otomoíes del Valle del Mezquital. 1. ed. México, D.F: CDI, Comisión Nacional para el Desarrollo de los Pueblos Indígenas (Pueblos indígenas del México contemporáneo). medRxiv preprint doi: https://doi.org/10.1101/2021.06.07.21258522; this version posted June 12, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC 4.0 International license .

Organization, W. H. and Fund (UNICEF), U. N. C. (2020a) Statement on the second meeting of the International Health Regulations (2005) Emergency Committee regarding the outbreak of novel coronavirus (2019-nCoV). Available at: https://www.who.int/news/item/30-01-2020-statement-on-the-second-meeting-of-the- international-health-regulations-(2005)-emergency-committee-regarding-the-outbreak-of- novel-coronavirus-(2019-ncov) (Accessed: 27 May 2021).

Organization, W. H. and Fund (UNICEF), U. N. C. (2020b) ‘Water, sanitation, hygiene, and waste management for the COVID-19 virus: interim guidance, 19 March 2020’. Available at: https://apps.who.int/iris/handle/10665/331499 (Accessed: 27 May 2021).

Pérez, C. J. P., Zacatenco, G. M. G. and Martínez, M. M. (2006) ‘Evolución del rendimiento y de la producción de maíz de riego en el Valle del Mezquital’, INIFAPSAGARPA. , Hidalgo. Publicación especial, (14).

Pérez Camarillo, J. P. (2002) ‘Carbón de la Espiga de Maíz. Descripción, síntomas y prevención.’ INIFAP-CIRCE-C.E. Pachuca. FHP. Available at: https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=&ved=2ahUKEwji 8fW-nPXwAhUHXK0KHdS- AX8QFjAAegQIAxAD&url=http%3A%2F%2Fpublico.senasica.gob.mx%2Fincludes%2F asp%2Fdownload.asp%3FIdDocumento%3D26871%26IdUrl%3D67966%26objeto%3DP %25E1gina%26IdObjetoBase%3D6073%26down%3Dtrue&usg=AOvVaw3KlZVvMjGFY FnRD0uzPzRG.

Randazzo, W. et al. (2020) ‘SARS-CoV-2 RNA in wastewater anticipated COVID-19 occurrence in a low prevalence area’, Water Research, 181, p. 115942. doi: 10.1016/j.watres.2020.115942.

Rossette, A. (2017) ‘En este pueblo mexicano todo lo que corre, se arrastra o vuela va a la cazuela’, The New York Times, 13 May. Available at: https://www.nytimes.com/es/2017/05/13/espanol/america-latina/feria-gastronomica- mexico-mezquital.html (Accessed: 11 May 2021).

SAGARPA, S. de A., Ganadería, Desarrollo Rural, Pesca y Alimentación. (2003) medRxiv preprint doi: https://doi.org/10.1101/2021.06.07.21258522; this version posted June 12, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC 4.0 International license .

‘Resultados del Estudio del Diagnóstico Sectorial’. Gobierno de Mexico. Available at: https://docplayer.es/10681820-Resultados-del-estudio-de-diagnostico-sectorial-en-el- estado-de-hidalgo-2010.html (Accessed: 31 May 2021).

Seth Denizen, K. (2021) Seth Denizen, ‘Thinking Through Soil: Case Study from the Mezquital Valley’, Harvard Graduate School of Design. Available at: https://www.gsd.harvard.edu/event/seth-denizen-thinking-through-soil-case-study-from- the-mezquital-valley/ (Accessed: 28 May 2021).

SIAP (2018) ‘Atlas Agroalimentario 2012 – 2018.’ SAGARPA. Available at: https://nube.siap.gob.mx/gobmx_publicaciones_siap/pag/2018/Atlas-Agroalimentario- 2018.

State of Hidalgo Government (2021) Available at: https://datamexico.org/es/profile/geo/hidalgo-hg.

Tang, B. et al. (2020) ‘An updated estimation of the risk of transmission of the novel coronavirus (2019-nCov)’, Infectious Disease Modelling, 5, pp. 248–255. doi: 10.1016/j.idm.2020.02.001.

Torkzaban, S. et al. (2006) ‘Role of air-water interfaces on retention of viruses under unsaturated conditions’, Water Resources Research, 42(12). doi: https://doi.org/10.1029/2006WR004904.

Vallejo, J. A. et al. (2020) ‘Predicting the number of people infected with SARS-COV-2 in a population using statistical models based on wastewater viral load’, medRxiv, p. 2020.07.02.20144865. doi: 10.1101/2020.07.02.20144865.

Wang, S. et al. (2020) ‘High-throughput wastewater analysis for substance use assessment in central New York during the COVID-19 pandemic’, Environmental Science: Processes & Impacts, 22(11), pp. 2147–2161. doi: 10.1039/D0EM00377H.

Wang, W. et al. (2020) ‘Detection of SARS-CoV-2 in Different Types of Clinical Specimens’, JAMA. doi: 10.1001/jama.2020.3786. medRxiv preprint doi: https://doi.org/10.1101/2021.06.07.21258522; this version posted June 12, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC 4.0 International license .

Wu, F. et al. (2020) ‘SARS-CoV-2 titers in wastewater foreshadow dynamics and clinical presentation of new COVID-19 cases’, medRxiv, p. 2020.06.15.20117747. doi: 10.1101/2020.06.15.20117747.

Zhang, X. et al. (2020) ‘Epidemiological, clinical characteristics of cases of SARS-CoV-2 infection with abnormal imaging findings’, International Journal of Infectious Diseases, 94, pp. 81–87. doi: 10.1016/j.ijid.2020.03.040.

medRxiv preprint doi: https://doi.org/10.1101/2021.06.07.21258522; this version posted June 12, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC 4.0 International license .

Supplementary material captions

A separate supplementary material PDF file is available.

Supplementary file 1. Code details, boundary conditions and description of the process for adjusting the transport model of virus in interfaces to an ordinary differential equation that describes the dispersion of SARS-CoV-2 in the Tula River. https://nbviewer.jupyter.org/github/yaxastro3000/COVID_CIAD_URH/blob/c65ac45af147 36023e94eb087aea8d541a7dac68/MODEL_COVID_CIAD_URH.ipynb

Supplementary tables

Supplementary Table S1. Soil samples collected for this study.

Supplementary Table S2. Ct values obtained from RT-qPCR detection of SARS-CoV-2 and PMMV in viral nucleic acids from soil and produce samples

Supplementary Table S3. Physicochemical parameters of water samples from Tula River (RW) and irrigation canals (CW)

Supplementary Table S4. Microbiological analyses of water and soil sample

Supplementary figures

Supplementary Figure S1. Correlations between SARS-CoV-2 and physicochemical variables (top) and between SARS-CoV-2 and microbiological variables (bottom) in water samples.

Supplementary Figure S2. Biodegradability index (COD/BOD) measured in Tula River, Tepeji River and Salado River, shows trends similar to those od SARS-CoV-2 concentration. Each label represents the sample location indicated in figure 1c.